Thinning algorithms for three-dimensional gray images and their application to medical images with comparative evaluation of performance

The spherical parameterization is important for the correspondence problem that is a major part of statistical shape modelling for the reconstruction of patient-specific 3D models from medical images. In this paper, we present comparative studies of five common spherical mapping methods applied to the femur and tibia models: the Issenburg et al. method, the Alexa method, the Saba et al. method, the Praun et al. method and the Shen et al. method. These methods are evaluated using three sets of measures: distortion property, geometric error and distance to standard landmarks. Results show that the Praun et al. method performs better than other methods while the Shen et al. method can be regarded as the most reliable one for providing an acceptable correspondence result. We suggest that the area preserving property can be used as a sufficient condition while the angle preserving property is not important when choosing a spherical mapping method for correspondence application.

Download Full-text

Open Source Applications for Image Visualization and Processing in Neuroimaging Training

Human-Computer Interaction ◽

10.4018/978-1-4666-8789-9.ch064 ◽

2015 ◽

pp. 1319-1332

Author(s):

Juan A. Juanes ◽

Pablo Ruisoto ◽

Alberto Prats-Galino ◽

Andrés Framiñán

Keyword(s):

Magnetic Resonance ◽

Medical Imaging ◽

Open Source ◽

Medical Images ◽

Three Dimensional ◽

Mr Images ◽

Image Visualization ◽

3D Slicer ◽

Volumetric Reconstruction ◽

Key Features

The aim of this paper is to demonstrate the major role and potential of three of the most powerful open source computerized tools for the advanced processing of medical images, in the study of neuroanatomy. DICOM images were acquired with radiodiagnostic equipment using 1.5 Tesla Magnetic Resonance (MR) images. Images were further processed using the following applications: first, OsiriXTM version 4.0 32 bits for OS; Second, 3D Slicer version 4.3; and finally, MRIcron, version 6. Advanced neuroimaging processing requires two key features: segmentation and three-dimensional or volumetric reconstruction. Examples of identification and reconstruction of some of the most complex neuroimaging elements such vascular ones and tractographies are included in this paper. The three selected applications represent some of the most versatile technologies within the field of medical imaging.

Download Full-text

Medical images segmentation based on improved three-dimensional pulse coupled neural network

International Journal of Wireless and Mobile Computing ◽

10.1504/ijwmc.2017.087358 ◽

2017 ◽

Vol 13 (1) ◽

pp. 72 ◽

Cited By ~ 1

Author(s):

Maxin Wang ◽

Xinzheng Xu ◽

Guanying Wang ◽

Shifei Ding ◽

Tongfeng Sun

Keyword(s):

Neural Network ◽

Medical Images ◽

Three Dimensional ◽

Pulse Coupled Neural Network

Download Full-text

A COMPARATIVE EVALUATION OF FEATURES FOR MEDICAL IMAGE MODALITY CLASSIFICATION

Jurnal Teknologi ◽

10.11113/jt.v78.9550 ◽

2016 ◽

Vol 78 (8-2) ◽

Author(s):

Sameer Ahmad Khan ◽

Suet Peng Yong ◽

Uzair Iqbal Janjua

Keyword(s):

Comparative Evaluation ◽

Medical Image ◽

Medical Images ◽

Image Categorization ◽

Discriminatory Ability ◽

Automatic Categorization ◽

Image Characteristic ◽

The Stability ◽

Content Information ◽

Image Modality

Medical images are increasing at an alarming rate. This increasing number of images affects the interpreting capacity of radiologists. In order to reduce the burden of radiologists, automatic categorization of medical images based on modality is the need of the hour. Because image modality is an important and fundamental image characteristic. The important factor in the automatic medical image categorization based on modality are the features used for categorization purpose, because nice treatment on these subtleties can lead to good results. Many descriptors have been proposed in the literature for medical image categorization. It is unclear which descriptor encodes the content information efficiently. The descriptors that are calculated from these medical images should be descriptive, distinctive and robust to various transformations. The stability of these descriptors are evaluated under various transformations and are then analyzed for their discriminatory ability for the task of classification. In this study the criteria of transformations, repeatability, matching and classification accuracy on the basis of precision recall is used to evaluate the performance of these descriptors. The experimental results illustrates that among global descriptors local features patches histogram and among local descriptors SIFT encodes the content information quite efficiently.

Download Full-text